There are several ways available that detect a focus state of a photographing lens in a digital camera that performs photographing using an image sensor. An apparatus disclosed in Japanese Patent Laid-Open No. 58-24105 is designed to perform focus state detection based on a pupil division method using a two-dimensional sensor having a microlens formed at each pixel on the sensor. According to the apparatus disclosed in Japanese Patent Laid-Open No. 58-24105, a photoelectric converter of each pixel of the image sensor is divided into a plurality of parts. The photoelectric converter that is divided in this manner is designed to receive light beams transmitted through different areas on a pupil of a photographing lens through a microlens.
Further, Japanese Patent No. 2959142 discloses a solid-state image sensing apparatus that also serves as an image sensor, in which pixels whose relative positions between a microlens and a photoelectric converter are shifted are arranged two-dimensionally. According to the solid-state image sensing apparatus disclosed in Japanese Patent No. 2959142, when detecting a focus state of a photographing lens, the focus state of the lens is detected based on images generated with pixel arrays that have different relative shift directions between a microlens and a photoelectric converter. In contrast, when sensing a normal image, adding pixels that correspond to each other and that have different relative shift directions between a microlens and a photoelectric converter generates an image.
The present applicant disclosed a solid-state image sensing apparatus that performs focus state detection based on a pupil division method using a CMOS-type image sensor (solid-state image sensing apparatus) that is used in a digital still camera in Japanese Patent Laid-Open No. 2005-106994. According to the solid-state image sensing apparatus disclosed in Japanese Patent Laid-Open No. 2005-106994, some pixels among a large number of pixels constituting the solid-state image sensing apparatus have a configuration in which a photoelectric converter for detecting a focus state of a photographing lens is divided into two sections. The photoelectric converter is designed to receive light beams transmitted through a predetermined area on a pupil of the photographing lens through a microlens.
FIG. 10 is an explanatory view concerning the distribution of received light of a pixel that performs focus state detection that is positioned at the center of the solid-state image sensing element disclosed in Japanese Patent Laid-Open No. 2005-106994, and shows areas on the pupil of the photographing lens which pass light that the two sections of the photoelectric converter divided into two parts can receive, respectively. The circle in the figure denotes an exit pupil of the photographing lens, and the white areas Sα and Sβ are areas which pass light, that the photoelectric converter that is divided into two parts, can receive. These areas are normally designed so as to be symmetric with respect to the optical axis of a photographing lens (point of intersection of x-axis and y-axis in the figure).
In the camera, a correlation operation is performed with respect to an image that is generated by light beams transmitted through the area Sα on the pupil of the photographing lens and an image that is generated by light beams transmitted through the area Sβ to detect a focus state of the photographing lens. A method that performs focus state detection by performing a correlation operation with respect to images generated by light beams transmitted through different pupil areas of a photographing lens is disclosed in Japanese Patent Laid-Open No. 5-127074.
In general, a CMOS-type solid-state image sensing element is manufactured by being subjected to multiple mask processes. Although the manufacturing is performed while alignment is performed between each mask process, misalignments occur between members manufactured at an initial stage and members manufactured at subsequent stages. That is, since a photoelectric converter of a solid-state image sensing element is formed at an initial stage of the manufacturing process and a microlens is formed at a final stage thereof, a misalignment often occurs between the photoelectric converter and the microlens.
FIG. 11A and FIG. 11B are views that illustrate the distribution of received light of a solid-state image sensing element on a pupil of a photographing lens. FIG. 11A shows the distribution on a light receiving area in a case in which there is a deviation from a designed value with respect to positions of photoelectric converters and a position of a microlens on a pixel that performs focus state detection in which a photoelectric converter is divided into two parts. In FIGS. 11A and 11B, a light receiving area Sα of one of the photoelectric converters and a light receiving area Sβ of the other photoelectric converter deviate in the −x direction with respect to the optical axis of the photographing lens (point of intersection of x-axis and y-axis in the figure).
When detecting a focus state of a photographing lens of a camera, normally focus state detection is possible not only with respect to a subject that is located in the center of the photographing screen, but also with respect to a subject located at the periphery of the photographing screen.
FIG. 11B is a view that describes the distribution of received light of a focus state detection pixel arranged at the periphery of a photographing screen, in a solid-state image sensing element in which the positions of photoelectric converters and the position of a microlens deviate from a designed value. Since vignetting of light beams by the lens frame of the photographing lens occurs at the periphery of the photographing screen, the light receiving area Sα corresponding to one of the photoelectric converters narrows as shown in FIG. 11B.
Consequently, the degree of agreement between an image generated by light beams transmitted through the area Sα on the pupil of the photographing lens and an image generated by light beams transmitted through the area Sβ decreases. As a result, even when a correlation operation is performed based on an image generated by light beams transmitted through the area Sα and an image generated by light beams transmitted through the area Sβ, there is the drawback that it is not possible to perform focus state detection with a high accuracy.
Further, since a distribution of received light on the pupil of the photographing lens of a pixel group forming a part of the solid-state image sensing apparatus which detects a focus state of the photographing lens has a band shape, the output from the pixel group tends to be affected by noise when a subject is dark or the contrast of the subject is small. Therefore, there is the drawback that it is not possible to perform focus state detection with a high accuracy.